E-Sports Training System Based on Intelligent Gesture Recognition.

Area of Science:

• Computer Science
• Human-Computer Interaction
• Sports Science

Background:

• Traditional e-sports training lacks objective performance metrics.
• Intelligent systems can provide real-time feedback for skill improvement.

Purpose of the Study:

• To develop an e-sports training system using intelligent gesture recognition.
• To evaluate player training effects via gesture analysis.
• To propose and validate an improved feature extraction algorithm.

Main Methods:

• Developed an e-sports training system integrating intelligent gesture recognition.
• Proposed an improved SLC-Harris feature extraction algorithm.
• Utilized KLT optical flow for feature point tracking.
• Calculated pure visual pose using epipolar geometry, triangulation, and PnP algorithms.

Main Results:

• The improved SLC-Harris algorithm's feasibility was verified on the EUROC dataset.
• The system successfully tracked feature points and calculated player poses.
• Experimental results indicate the system has a positive effect on e-sports training.

Conclusions:

• The proposed intelligent gesture recognition system offers a viable method for e-sports training.
• The improved feature extraction algorithm enhances system performance.
• This approach provides objective feedback for player development.